Maxwell-Boltzmann speed distribution — derivation (key for propulsion): Question bank

Before we begin, one reminder of the players so every symbol below is earned:

$v$ = speed (a length per time, always $\geq 0$ — no direction).
$v_{x}, v_{y}, v_{z}$ = the three velocity components: a molecule's velocity is an arrow in space, and $v_{x}$ is how fast it moves along the $x$ -direction (it can be negative — leftward — unlike speed). The speed rebuilds from them via $v = v_{x}^{2} + v_{y}^{2} + v_{z}^{2}$ .
$T$ = temperature, $m$ = mass of one molecule, $k_{B}$ = Boltzmann's constant (the number that converts temperature into energy per molecule).
$β$ = the shorthand constant $\frac{m}{2 k _{B} T}$ that sits in the exponent $e^{- β v^{2}}$ ; it measures how sharply the distribution penalises high speeds (large $β$ = cold/heavy = steep drop-off).
$v_{p} = 2 k_{B} T / m$ (peak), $\overset{v}{ˉ} = 8 k_{B} T / π m$ (mean), $v_{r m s} = 3 k_{B} T / m$ (root-mean-square).

The picture every question below points back to — memorise its shape:

Figure — Maxwell-Boltzmann speed distribution — derivation (key for propulsion)

True or false — justify

True or false: If temperature is fixed, every molecule in the gas moves at

v_{r m s}

False —

v_{r m s}

is just one summary number of a whole spread; molecules occupy speeds from near

0

to arbitrarily large, only weighted by

f (v)

True or false:

f (v)

is a symmetric bell curve like a Gaussian.

False — the

v^{2}

factor pushes the curve to zero at

v = 0

and stretches a long tail to the right, so it is skewed, not symmetric.

True or false: The area under

f (v)

changes when you heat the gas.

False — the area is always

1

(it must, since some speed is certain); heating widens and lowers the curve but conserves total area.

True or false: Doubling

T

doubles

v_{p}

False —

v_{p} \propto T

, so doubling

T

multiplies

v_{p}

by only

2 \approx 1.41

True or false: A lighter gas at the same temperature has a broader, faster distribution.

True — all speeds scale as

1/ m

, so smaller

m

shifts and spreads the whole curve toward higher speed (why H₂ is a prized propellant, see Rocket Propulsion & Specific Impulse).

True or false: The most probable speed

v_{p}

equals the average speed

\overset{v}{ˉ}

False — the skew drags the mean above the peak, giving the fixed order

v_{p} < \overset{v}{ˉ} < v_{r m s}

True or false: At the same temperature, two gases with equal molecular mass have identical speed distributions.

True —

f (v)

depends only on

m

and

T

(through

k_{B} T / m

), not on chemistry, so equal

m

at equal

T

gives an identical curve.

True or false: The velocity component

v_{x}

has the same distribution shape as the speed

v

False — each component

v_{x}

is a symmetric Gaussian (peaked at

0

, can be negative), while the speed is one-sided and peaks away from

0

because of the

4 π v^{2}

shell factor.

True or false:

⟨ v^{2} ⟩ = ⟨ v ⟩^{2}

False — the mean of the squares exceeds the square of the mean by the variance, which is strictly positive for any real spread of speeds.

Spot the error

"The peak of

f (v)

tells you the average speed of the gas." — find the flaw.

The peak is

v_{p}

, the most common speed; because the curve is skewed right, the true average

\overset{v}{ˉ}

sits noticeably above the peak.

f (v) = (\frac{m}{2 π k _{B} T})^{3/2} e^{- m v^{2} /2 k_{B} T}

." — what is missing?

The

4 π v^{2}

spherical-shell factor is dropped; this expression is the density per velocity vector, not per speed, and it wrongly peaks at

v = 0

(compare the correct form in the reference callout).

"Since

e^{- m v^{2} /2 k_{B} T}

is largest at

v = 0

, most molecules are nearly at rest." — flaw?

The exponential is largest at

0

, but the

v^{2}

factor is zero there, so

f (0) = 0

; almost no molecules are truly at rest.

"Kinetic energy of the gas uses the mean speed, so

KE \propto \overset{v}{ˉ}^{2}

." — flaw?

Energy needs

⟨ v^{2} ⟩

, not

⟨ v ⟩^{2}

; using

\overset{v}{ˉ}^{2}

undercounts because fast molecules carry disproportionate energy.

"To find

v_{p}

we set

\frac{d}{d v} (e^{- m v^{2} /2 k_{B} T}) = 0

." — flaw?

You must differentiate the whole

v^{2} e^{- m v^{2} /2 k_{B} T}

; the exponential alone has its extremum at

v = 0

, which is the wrong (minimum) point.

"Because space is isotropic, the speed distribution has no preferred value, so it is flat." — flaw?

Isotropy removes preferred directions, not preferred speeds; the energy penalty

e^{- m v^{2} /2 k_{B} T}

still makes some speeds far more likely than others.

"Raising temperature raises the peak height of

f (v)

because molecules move faster." — flaw?

Faster molecules spread the curve wider, and since the area is fixed at

1

, the peak height actually falls as

T

rises.

Why questions

Why must a gas have a distribution of speeds rather than one shared speed?

Constant collisions randomly trade energy between molecules, so any single-speed state is instantly destroyed; only a statistical spread is stable (Kinetic Theory of Gases).

Why does each velocity component come out as an exponential of a quadratic,

e^{- β v_{x}^{2}}

Three facts chain together — (1) independence of

v_{x}, v_{y}, v_{z}

makes the joint distribution a product

g (v_{x}) g (v_{y}) g (v_{z})

; (2) isotropy demands that product depend only on

v_{x}^{2} + v_{y}^{2} + v_{z}^{2}

; (3) the only function that turns a product into a function of a sum-of-squares is the exponential, since

e^{a} e^{b} e^{c} = e^{a + b + c}

, forcing each exponent to be

\propto v_{x}^{2}

and negative to stay finite.

Why does the constant

β = \frac{m}{2 k _{B} T}

turn out that way?

The Equipartition Theorem fixes the mean energy per component at

\frac{1}{2} k_{B} T

, which forces

⟨ v_{x}^{2} ⟩ = 1/2 β = k_{B} T / m

and hence

β = m /2 k_{B} T

Why does the speed distribution carry a

4 π v^{2}

factor?

All velocity vectors of a given speed lie on a sphere of radius

v

in velocity space, and the number of such states scales with that sphere's surface area

4 π v^{2}

Why is the high-speed tail the part that matters most for propulsion and reactions?

The fastest molecules dominate escape, reaction rates, and effusion; a small temperature rise fattens this tail dramatically even if the peak barely moves (Effusion and Graham's Law).

Why does hydrogen give a higher exhaust speed than CO₂ at the same temperature?

Every characteristic speed scales as

1/ m

, so H₂'s tiny mass pushes its whole distribution to much higher speeds, raising specific impulse.

Why is the ratio

v_{p} : \overset{v}{ˉ} : v_{r m s}

the same for every gas at every temperature?

Each speed is a pure number times

k_{B} T / m

, and that common factor cancels in the ratio, leaving the universal

2 : 8/ π : 3

Why does the Boltzmann Distribution

e^{- E / k_{B} T}

sit "behind" the Maxwell–Boltzmann speed law?

Substituting the kinetic energy

E = \frac{1}{2} m v^{2}

into the general Boltzmann weight

e^{- E / k_{B} T}

produces exactly the exponential factor in

f (v)

Edge cases

What is

f (v)

exactly at

v = 0

Exactly

0

, because the

v^{2}

shell factor vanishes there — no molecule has literally zero speed with finite probability density.

What happens to

f (v)

v \to \infty

It decays to

0

, because the exponential

e^{- m v^{2} /2 k_{B} T}

crushes the polynomial

v^{2}

growth for large speeds.

In the limit

T \to 0

, what does the distribution look like?

It collapses toward a spike at

v = 0

— the exponential penalty becomes infinitely steep, so almost all molecules freeze to near-zero speed.

In the limit

T \to \infty

, what happens to the three characteristic speeds?

All of them (

v_{p}, \overset{v}{ˉ}, v_{r m s}

) grow without bound as

T

, and the curve flattens and stretches indefinitely to the right.

If the molecular mass

m \to \infty

at fixed

T

, where does the peak go?

Toward

v = 0

, since

v_{p} \propto 1/ m

— very heavy molecules are sluggish and cluster at low speed.

Does a single molecule "have" a Maxwell–Boltzmann distribution at one instant?

No —

f (v)

describes the ensemble of many molecules (or one molecule sampled over long time); at one instant a single molecule just has one definite speed.

f (v)

ever negative for some speed?

Never — it is a product of

v^{2} \geq 0

, a positive exponential, and a positive constant, so it is non-negative everywhere, as any probability density must be.

Connections

Kinetic Theory of Gases — why collisions force a statistical spread
Equipartition Theorem — fixes $β = m /2 k_{B} T$
Boltzmann Distribution — the $e^{- E / k_{B} T}$ ancestor of $f (v)$
Mean Free Path — uses $\overset{v}{ˉ}$
Rocket Propulsion & Specific Impulse — the low- $m$ , high-speed payoff
Effusion and Graham's Law — tail- and $\overset{v}{ˉ}$ -driven escape rates